Mathematical Modeling of Option Pricing with an Extended Black-Scholes Framework
Nikhil Shivakumar Nayak · 2025
Read Mathematical Modeling of Option Pricing with an Extended Black-Scholes Framework by Nikhil Shivakumar Nayak (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-03175.txt.
Introduction
Nikhil Shivakumar Nayak, writing alone in 2025, works on pricing financial options with an extended classic model. The Black-Scholes model prices options with a fixed view of risk. Real markets show shifting volatility and interest rates. Nayak builds both shifts into the model's core equation.
He solves the equation with a grid method on a computer. He also builds a learning model of a different kind for comparison. Both priced options on Google stock and were checked against past market data. The learning model was more accurate. The grid method was faster to run. The study suggests that joining the two styles could give pricing that is both sharp and cheap to compute.
Catalog record
| CREATOR | Nikhil Shivakumar Nayak |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 19 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2504-03175
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Mathematical Modeling of Option Pricing with an Extended Black-Scholes Framework Nikhil Shivakumar Nayak Abstract This study investigates enhancing option pricing by extending the Black-Scholes model to include stochastic volatility and interest rate variability within the Partial Differential Equation (PDE). The PDE is solved using the finite difference method. The extended Black-Scholes model and a machine learning-based LSTM model are developed and evaluated for pricing Google stock options. Both models were backtested using historical market data. While the LSTM model exhibited higher predictive accuracy, the finite difference method demonstrated superior computational efficiency. This work provides insights into model performance under varying market conditions and emphasizes the potential of hybrid approaches for robust financial modeling. 1 Introduction The Black-Scholes model [ 7 ] , formulated by Fischer Black, Myron Scholes, and Robert Merton, revolutionized financial mathematics by providing a closed-form solution for European option pricing. Its equation rests on the assumption of constant volatility and a risk-free interest rate, offering a framework that simplified risk management and derivatives pricing. This model laid the foundation for financial engineering and led to the rapid expansion of derivative markets. However, the model has notable limitations. It assumes constant volatility and interest rates, disregarding real-world market dynamics where volatility often fluctuates and interest rates vary over time. Moreover, Black-Scholes [ 7 ] assumes continuous trading and fails to capture drastic market movements or the volatility smile effect. These simplifications can lead to significant inaccuracies in estimating the fair value of options. To address these shortcomings, researchers have proposed variants of the Black-Scholes model [ 7 ] that incorporate more nuanced factors affecting option pricing. Models such as the Heston model [ 10 ] and the Cox-Ingersoll-Ross (CIR) process [ 9 ] aim to better represent the realities of market behavior. In particular, extending the Black-Scholes model to account for volatility and interest rate variability is essential for robust option pricing, which is the focus of this work. Such enhancements provide investors and financial institutions with more accurate assessments of market risks and opportunities. In parallel with these analytical approaches, machine learning models like Long Short-Term Memory (LSTM) networks [ 4 ] have gained traction in financial applications due to their ability to handle complex temporal dependencies. LSTM networks, a variant of recurrent neural networks (RNNs), are well-suited for time series forecasting because they can capture long-term patterns while minimizing the impact of vanishing gradients. By utilizing LSTM models for option pricing, we can potentially achieve better predictive accuracy and adaptability. A good model for option pricing should effectively balance accuracy and computational efficiency, while maintaining robustness under varying market conditions. This project aims to compare an enhanced Black-Scholes model, solved using finite difference methods [ 5 ] , with an LSTM-based machine learning model for pricing options. Through rigorous backtesting and evaluation, we intend to understand the strengths and limitations of each approach. 2 Model Description In this section, we will discuss the mathematical models employed in this study for option pricing. Each model relies on inputs such as stock prices, volatility, and interest rates to forecast the option price. 2.1 Extended Black-Scholes PDE Model Derivation of the Extended Black-Scholes Equation In financial markets, the Black-Scholes equation is crucial for pricing options by assuming a constant volatility and a constant risk-free interest rate. However, to make the model more realistic, we extend it to include stochastic volatility and interest rate variability. The original Black-Scholes equation [ 7 ] is given by: \frac{\partial V}{\partial t}+\frac{1}{2}\sigma^{2}S^{2}\frac{\partial^{2}V}{\partial S^{2}}+rS\frac{\partial V}{\partial S}-rV=0 where V(t,S) is the price of the option, S is the stock price, t is time, \sigma is the volatility of the stock (constant in the original model), r is the risk-free interest rate (constant in the original model). To incorporate stochastic volatility, let’s denote the volatility as a stochastic process \sigma(t) . The instantaneous variance follows another stochastic process, often modeled as a mean-reverting process like in the Heston model [ 10 ] : d\sigma^{2}=\kappa(\theta-\sigma^{2})dt+\xi\sqrt{\sigma^{2}}dW_{\sigma} where \kappa , \theta , and \xi are model parameters, and W_{\sigma} is a Brownian motion [ 8 ] that may be correlated with the stock price’s Brownian motion. Similarly, to incorporate interest rate variability, we assume the risk-free rate r is also a stochastic process. A common model for this is the Vasicek model [ 6 ] : dr=a(b-r)dt+sdW_{r} where a , b , and s are parameters, and W_{r} is another Brownian motion that can be correlated with W and W_{\sigma} . Incorporating both stochastic volatility and interest rate variability into the Black-Scholes framework [ 7 ] not only changes the dynamic terms related to S and t , but also introduces dependencies on \sigma and r . This leads to a more complex partial differential equation (PDE) that now includes partial derivatives with respect to these stochastic components. The revised equation incorporating stochastic volatility and interest rate variability is: \frac{\partial V}{\partial t}+\frac{1}{2}\sigma(t)^{2}S^{2}\frac{\partial^{2}V}{\partial S^{2}}+r(t)S\frac{\partial V}{\partial S}-r(t)V+\frac{\partial V}{\partial\sigma}\frac{d\sigma}{dt}+\frac{\partial V}{\partial r}\frac{dr}{dt}=0 The term \frac{\partial V}{\partial\sigma}\frac{d\sigma}{dt} accounts for the sensitivity of the option value to changes in volatility, where \frac{d\sigma}{dt} reflects the stochastic nature of volatility. The term \frac{\partial V}{\partial r}\frac{dr}{dt} captures the effect of stochastic interest rates on the option value, with \frac{dr}{dt} being the dynamic change in the interest rate. Stochastic Models for \sigma and r : Volatility (Heston Model [ 10 ] ): d\sigma^{2}=\kappa(\theta-\sigma^{2})dt+\xi\sqrt{\sigma^{2}}dW_{\sigma} Interest Rate (Vasicek Model [ 6 ] ): dr=a(b-r)dt+sdW_{r} To solve this complex PDE, simplifying assumptions and numerical methods are typically employed. The partial derivatives with respect to \sigma and r can be challenging to handle analytically and often require Monte Carlo simulations or finite difference methods for numerical solutions. Solving the Extended Black-Scholes Equation Using Finite Difference Method Given the complexity introduced by stochastic volatility and varying interest rates, the extended Black-Scholes equation we aim to solve is: \frac{\partial V}{\partial t}+\frac{1}{2}\sigma(t)^{2}S^{2}\frac{\partial^{2}V}{\partial S^{2}}+r(t)S\frac{\partial V}{\partial S}-r(t)V+\frac{\partial V}{\partial\sigma}\frac{d\sigma}{dt}+\frac{\partial V}{\partial r}\frac{dr}{dt}=0 To make this equation tractable for the finite difference method [ 5 ] , we make few assumptions. First, volatility \sigma and interest rate r processes are discretized over a grid. At each time step, these are approximated as constant, but can change from one step to the next. Next, the Brownian motions driving \sigma and r are assumed uncorrelated with each other and with the stock price’s Brownian motion. Finally, using Implicit and Explicit Finite Difference schemes, the option value at the next time step is calculated from known values at the current time step. Finite Difference Method Discretize the time t , stock price S , volatility \sigma , and interest rate r into a grid: t_{i}=i\Delta t for i=0,1,\ldots,N ; S_{j}=j\Delta S for j=0,1,\ldots,M ; \sigma_{k} and r_{l} are discretized over their respective ranges. The partial derivatives in the PDE can be approximated as follows: Time Derivative: \frac{\partial V}{\partial t}\approx\frac{V^{i+1}_{j,k,l}-V^{i}_{j,k,l}}{\Delta t} Stock Price Derivatives: \frac{\partial V}{\partial S}\approx\frac{V^{i}_{j+1,k,l}-V^{i}_{j-1,k,l}}{2\Delta S} \frac{\partial^{2}V}{\partial S^{2}}\approx\frac{V^{i}_{j+1,k,l}-2V^{i}_{j,k,l}+V^{i}_{j-1,k,l}}{\Delta S^{2}} Volatility and Interest Rate Derivatives are approximated similarly. Using this scheme, the value at the next time step is calculated by: V^{i+1}_{j,k,l}=V^{i}_{j,k,l}+\Delta t\left(-\frac{1}{2}\sigma_{k}^{2}S_{j}^{2}\frac{\partial^{2}V}{\partial S^{2}}-r_{l}S_{j}\frac{\partial V}{\partial S}+r_{l}V^{i}_{j,k,l}-\frac{\partial V}{\partial\sigma}\frac{d\sigma}{dt}-\frac{\partial V}{\partial r}\frac{dr}{dt}\right) Boundary Conditions: At S=0 the value of an option is zero as the stock value is zero. As S\to\infty the option value must be capped or grow linearly depending on the type of option (put or